sprint
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user which is subsequently processed by automated sub-agents, creating a surface for indirect prompt injection.\n
- Ingestion points: User-supplied arguments such as
name,features, andreasonfor approval are stored in.bkit/state/sprints/and documentation files.\n - Boundary markers: There are no explicit instructions or delimiters shown to separate user-provided strings from agent instructions in the provided templates.\n
- Capability inventory: The skill utilizes high-privilege tools including
Bash,Write,Edit, and the ability to spawn sub-agents (bkit:sprint-master-planner,bkit:sprint-report-writer).\n - Sanitization: No evidence of sanitization or validation of the content of user-provided strings before they are interpolated into prompts for sub-agents.\n- [COMMAND_EXECUTION]: The skill architecture relies on the execution of shell commands and Node.js scripts to perform its core functions.\n
- The
allowed-toolslist includesBash, which is used for sprint orchestration and lifecycle management.\n - The documentation describes the invocation of
node scripts/sprint-handler.jsand the injection of adeps.agentTaskRunnerwhich facilitates the execution of sub-agent tasks.
Audit Metadata